Billing Spike from 195h to 1298h in <24h for T4 Space (Impossible Usage)

Hi Hugging Face Team,

I’m facing a serious billing issue with one of my Spaces and would appreciate urgent assistance.

Space Details:

• Space Type: GPU Space

• Hardware: T4 Small

Observed Usage:

Up until yesterday:

• Runtime: 195 hours 32 minutes

• Cost: $78.22

Today:

• Runtime: 1298 hours 2 minutes

• Cost: $519.24

Issue:

There is a sudden increase of ~1100 hours within less than 24 hours , which is technically impossible.

Even with continuous GPU usage:

• Maximum possible = 24 hours/day per instance

• This spike would imply dozens of parallel instances, which is not the case.

Additional Context:

• I restarted the Space yesterday and it was functioning normally.

• Today morning, I observed the Space in a runtime error state.

• It’s possible the GPU remained allocated unintentionally, but even then, such a large spike is not feasible.

Concern:

This appears to be:

• A billing/metering inconsistency

• Or a delayed usage reconciliation issue

• Or potential duplicate counting of runtime

This is a significant and unexpected charge, so I would appreciate a prompt review.

Thank you.